]> git.itanic.dy.fi Git - linux-stable/commit
KVM: nSVM: hyper-v: Enable L2 TLB flush
authorVitaly Kuznetsov <vkuznets@redhat.com>
Tue, 1 Nov 2022 14:54:06 +0000 (15:54 +0100)
committerPaolo Bonzini <pbonzini@redhat.com>
Fri, 18 Nov 2022 17:59:18 +0000 (12:59 -0500)
commit3f4a812edf5cb0a50e65fbdfafdb3e688da18f16
tree2768590a78f023064ee3a3f8abdd640c7eb061c7
parentb415d8d417bbe5403626b74e1041101ac23d602f
KVM: nSVM: hyper-v: Enable L2 TLB flush

Implement Hyper-V L2 TLB flush for nSVM. The feature needs to be enabled
both in extended 'nested controls' in VMCB and VP assist page.
According to Hyper-V TLFS, synthetic vmexit to L1 is performed with
- HV_SVM_EXITCODE_ENL exit_code.
- HV_SVM_ENL_EXITCODE_TRAP_AFTER_FLUSH exit_info_1.

Note: VP assist page is cached in 'struct kvm_vcpu_hv' so
recalc_intercepts() doesn't need to read from guest's memory. KVM
needs to update the case upon each VMRUN and after svm_set_nested_state
(svm_get_nested_state_pages()) to handle the case when the guest got
migrated while L2 was running.

Reviewed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20221101145426.251680-29-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
arch/x86/include/asm/hyperv-tlfs.h
arch/x86/kvm/hyperv.h
arch/x86/kvm/svm/hyperv.c
arch/x86/kvm/svm/hyperv.h
arch/x86/kvm/svm/nested.c